Beyond Prompts: Dynamic Conversational Benchmarking of Large Language Models

TL;DR

This paper presents a dynamic benchmarking system for conversational AI that evaluates long-term memory, continual learning, and information integration through simulated multi-task interactions, revealing limitations of current large language models.

Contribution

It introduces a novel dynamic benchmarking framework that simulates realistic multi-task conversations to assess LLMs' long-term capabilities and highlights their challenges in interleaved task scenarios.

Findings

LLMs perform well on single tasks but struggle with interleaved tasks.

Short-context LLMs with LTM can match or outperform larger-context models.

Current benchmarks do not fully capture challenges in natural, multi-task interactions.

Abstract

We introduce a dynamic benchmarking system for conversational agents that evaluates their performance through a single, simulated, and lengthy user$\leftrightarrow$agent interaction. The interaction is a conversation between the user and agent, where multiple tasks are introduced and then undertaken concurrently. We context switch regularly to interleave the tasks, which constructs a realistic testing scenario in which we assess the Long-Term Memory, Continual Learning, and Information Integration capabilities of the agents. Results from both proprietary and open-source Large-Language Models show that LLMs in general perform well on single-task interactions, but they struggle on the same tasks when they are interleaved. Notably, short-context LLMs supplemented with an LTM system perform as well as or better than those with larger contexts. Our benchmark suggests that there are other challenges for LLMs responding to more natural interactions that contemporary benchmarks have heretofore not been able to capture.